Building the “Detail SAM”:
Before, we begin discussing the energy adjustments that were made on the “Detail SAM”
to get to the Final SAM, we should discuss how the “Detail SAM” was built in the first place.
There are few datasets that were used to build the “Detail SAM” worksheet. These datasets are
paste values from other sources to the “SAM 5 Nov 04” Workbook. The ones of interest are from
the Household Commodity Demand workbook and the V0304.xls workbook (I don’t have either
of them):
1)
The matrix with the transactions between the households and the commodity
sectors (i.e. how much money each household category pays each commodity
sector) is obtained from the Aggregate Household Consumption worksheet of
the Household Commodity Demand workbook. This matrix enters as paste
values in the “HH Agg Cons Raw” worksheet; however, for the household
accounts to balance (i.e. the sum of a row be equal to the sum of the respective
column) the total consumption of each household category needs to change.
However, the total household consumption needs to stay the same as in the
original data (i.e. 862.194). All these adjustments occur in the “HH Cons”
worksheet. From there we paste the “new” matrix to the “Detail SAM”.
2)
The matrix with the transactions between the industry sectors and the
commodity sectors (i.e. how much money each commodity sector pays each
industry sector) is obtained from the Aggregate Industry Consumption
worksheet from the CA Household Commodity Demand workbook. This enters
as paste values in the “Indus Prod” worksheet and from there as paste values in
the “detail SAM”.
3)
The matrix with the transactions among the industry sectors (i.e. the upper left
matrix of a SAM) is obtained from V0304.xls. This enters as paste values in
the “V0304” worksheet and from there it enters as paste values in the “Detail
SAM”.
IMPORTANT!!!!!!!
The “Detail
SAM” does not balance.
In particular, even though the household and commodity sectors balance (i.e. the sum of the row
is equal to the sum of the column for each sector), the industry sectors do not balance (e.g. the
sum of the row of OILREF is not equal to the sum of the column of OILREF).
From “Detail SAM” to Final SAM by Adjusting Energy:
First of all let’s see conceptually what happens in the worksheet “Energy” step by step:
1) There are some energy data (it can be new prices or even data on quantities) and you
want to update your SAM based on them.
2) Get the rows of the 4 energy accounts from the existing SAM (i.e. OILGAS, OILREF,
DISTEL, and DSTGAS) up until the commodity sectors and adjust them based on your
energy data (e.g. assume that you know that the real price increase of OILGAS is 10%;
therefore, you multiply all the cells in the row by 1.1 since OILGAS now is more
expensive).
3) You want to keep total household consumption constant (i.e. 862.194) no matter what
(this is given from an “exogenous” source), which in turn implies that you must keep the
value of all the commodity sectors in total constant (i.e. 862.194) since only households
buy commodities.
4) Focus on the transactions between commodities and the energy sectors. For simplicity
assume all the energy sources become more expensive. So, commodities have to pay
more for energy; however, for the total value to remain constant, we should “make” the
commodities pay less for all other inputs such that the total value remains the same (i.e.
862.194). That is, we should adjust the commodity-industry sectors matrix (i.e. the
second matrix in the worksheet).
5) Even though we kept the total commodity value constant, we have probably changed the
column totals of each of the commodity sectors. To fix this we need to adjust the rows
which correspond to the commodities in order to get the same totals as the column totals.
That happens by adjusting the household-commodity matrix (i.e. the third matrix in the
worksheet).
Now let’s see mechanically, one matrix at a time what happens:
First Matrix of the Worksheet (4 Rows of Energy Accounts up until Commodities)
First of all, from the “Energy Data.xls” we have two kinds of data (just ignore for now the
conversion coefficients): energy price levels associated with the 4 energy accounts for 1997 and
2003 (and of course we can obtain the real price increase from 1997 to 2003 since we also have
the wage levels) and we have quantity level data for five transactions.
Since five transactions is a small number compared to the total number of transactions,
we will first see the general formula for each row and then we will see how the value of each of
these five transactions is calculated.
Row of OILGAS:
Denote Old Transaction as the transaction from the “detail SAM”. We need the real prices
increases associated with this energy sector (i.e. Crude oil, Gasoline and Natural Gas Citygate);
we get those from “Energy Data.xls”. Since we have three prices associated with OILGAS we
will use the average.
Real Price Increase of OILGAS = Average (Real Price Increase of Crude oil, Real Price Increase
of Gasoline and Real Price Increase of Natural Gas Citygate)
So the formula is:
New Transaction = Old Transaction * (Real Price Increase of OILGAS)
Row of OILREF:
New Transaction = Old Transaction * (Real Price Increase of Gasoline)
Row of DISTEL:
From AGRIC up until MSCMFG:
New Transaction = Old Transaction * (Real Price Increase of Elec. Industrial)
From WHLDUR up until COTHR:
New Transaction = Old Transaction * (Real Price Increase of Elec. Commercial)
Row of DSTGAS:
New Transaction = Old Transaction * (Real Price Increase of Nat gas Commercial)
Five Exceptions:
Now, as it can be seen the old transactions are just adjusted based on prices; these
transactions come from the “V0304.xls” and the CA Household Commodity Demand workbook.
However, we have five transactions on quantities based on a different dataset (i.e. Energy
Data.xls) which is “perceived to be better than the above”. So we would like to incorporate these
new data in the new SAM; hence, these five transactions do not follow similar formulas as
above. Instead, they are formed as follows:
OILGAS  DISTEL: (Quantity) * (Conversion Coefficient) *(2003 PG&E procurement cost of
gas)/ (10^9)
OILGAS  DSTGAS: (Quantity) * (Conversion Coefficient) *(2003 PG&E procurement cost
of gas)/ (10^9)
OILGAS  OILREF: (Quantity) * (Conversion Coefficient) *(2003 Price of Crude Oil)/ (10^9)
DISTEL  CFUEL: (Quantity) * (Conversion Coefficient) *(2003 Price of Elec Residential)/
(10^9)
DSTGAS  CFUEL: (Quantity) * (Conversion Coefficient) *(2003 Price Natural Gas
Residential/ (10^9)
Notice that we divide by 10^9, since we want the transactions to be expressed in billions of
dollars.
Second Matrix of the Worksheet (Commodities-Industries Matrix)
We need to establish a few things before getting there:
1) SUM(all the energy adjusted transactions between Energy accounts and Commodities)Sum(all the transactions between Energy accounts and Commodities from the “detail
SAM”) = -4.98
2) Total Household Consumption (862.194) - Sum(all the transactions between Energy
accounts and Commodities from the “detail SAM”) = 829.45
So, the formula is:
New Transaction  Old Transaction -
(-4.98) * Old Transaction
829.45
Note that this formula is for all transactions, except the transactions between the commodities
and the 4 energy sectors; these are just paste values from the first matrix. (the OILGAS row is not
pasted from the first matrix)
Third Matrix of the Worksheet (Households-Commodities Matrix)
Note:
1) From the second matrix, we get new column totals for each commodity sector.
2) From “Indus Prod”, we have the old column totals for each commodity sector.
3) From “HH cons”, we have the old transactions between commodities and households.
4) From “HH Cons”, we also have the old row totals of household consumption.
So, the formula is:
New ij th Transaction  Old ij th Transaction 
Old Total Consumption of i th Hous
1

th
Total Hous Consumption
( New Column Total of j Comm  Old Column Total of j th Comm)
last part of equation is different from the excel formula
Final Step:
Just plug these 3 new matrices in the “detail SAM” to get the final SAM.